# Read experiment descriptions in Book Isle: separate factors, responses and uncontrolled conditions Language: en Canonical: [Read experiment descriptions in Book Isle: separate factors, responses and uncontrolled conditions]() 中文：[中文 HTML]() English: [English HTML]() Published: 2026-11-05T00:00:00+08:00 Modified: 2026-11-05T09:38:59+08:00 Author: Bison Use a fictional paper-glider passage to record the question, factor levels, response measure and missing controls without turning observed association into causation. English translation of the [Chinese original](), translated on 2026-11-05. Original publication dates, attribution, code and version notes are retained. Experiment chapters often compress materials, procedures and results into one paragraph, making intended changes easy to confuse with measured outcomes. Use Book Isle to retain the source location and make a manual experiment record. The example is a fictional paper-glider comparison, requiring no real throws and supporting no performance conclusion about a design. Separate question, manipulation, measurement and limits. The app supports reading and notes; it does not automatically assess study quality or establish causation. ## Write a question traceable to the passage Choose an accessible experimental description, record title, edition, chapter, example number and a distinctive phrase, and bookmark near its question. State what the author compares in one sentence before extracting conclusions. For the fictional example: does gliding distance differ between two paper thicknesses under the same launching setup? Keep later fields tied to that question. If the text only reports past observations without deliberately changing conditions, label it observational rather than rewriting it as an intervention. A vague question should be marked unclear, not silently perfected through imagination. Search helps return to the passage; it does not supply the author’s missing objective. ## Record factor levels and units Record the intended input as paper thickness, with provisional levels of 0.10 mm and 0.20 mm explicitly labelled as invented passage values. Distinguish thickness from mass and area rather than treating all paper properties as interchangeable. If the book gives grammage, record that instead of silently calling it thickness. Attach each level to its source sentence. If the author changes both wing shape and paper type, record both changes instead of attributing the difference to only one. Leave missing names, levels or units unresolved rather than writing zero. This is a reading record, not advice to build or fly an object using these values. ## Specify how the response is measured Define the response as horizontal glide distance in metres and specify its endpoints. The fictional passage can measure from the vertical projection of release to first ground contact, not treat airborne path length and horizontal distance as the same quantity. If it only says flew farther, flag a missing measurement definition and look in context or an appendix. A response is neither an aspiration to succeed nor another name for the input factor. If a table also gives flight time, record it separately instead of averaging seconds together with metres. Find the measurement wording first, then paraphrase it; groups cannot be meaningfully compared until the response is clear. ## Separate stated controls from unknown conditions List explicitly maintained conditions, such as sheet area, folding method and release height, attaching evidence to each. If wind, launcher or measuring tool is not mentioned, write not reported rather than assuming identical conditions from the phrase same setup. Keep controls separate from factor levels; two thickness levels are not themselves a constant condition. If something changed without control, note a possible influence on comparison without asserting that it caused the result. Use a short note or paper table, not an assumed built-in experiment form in Book Isle. Separate explicit text, reader inference and unknowns so a tidy record cannot create false completeness. ## Distinguish repetitions, order and groups If the text says three trials at each level, record three measurements in each of two groups, not three overall. Check whether order is alternating, random or unreported: repetition does not automatically mean random assignment. If only group means are provided, flag missing individual observations instead of copying a mean three times as raw data. Leave empty cells in the fictional record rather than inventing realistic-looking results for demonstration. Identify whether the sample-count unit is gliders, throws or participants, without adding different levels together. This checks completeness of the design description, not statistical significance or a reliable performance difference. ## Separate reported results from causal wording Return to the results paragraph and record what was observed separately from the author’s explanation. A larger group mean should remain that comparison, not become thicker paper necessarily flies farther. Verify whether there was intervention, reasonable control of other conditions, described allocation and defined measurement; this record alone cannot adjudicate causation. Preserve qualifications such as associated, may and could help instead of compressing them into proves. If the author states limitations, retain their location with the result rather than extracting only the strongest sentence. The purpose is to make a conclusion’s boundaries visible, not to certify a short passage as scientific. ## Use missing fields to guide rechecking Hide personal interpretation and point each field—question, input, response, controls and count—back to a source sentence. For a missing field, search adjacent chapters or appendices, retaining not reported if it remains absent instead of placing your supplement in the author’s column. If table headings conflict with prose, record both locations and check for a different experiment or unit. Inspect the original page visually when copied table columns shift; search text alone is insufficient. Reconstruct the comparison from your record. If you still cannot say what was measured, return to the response definition rather than adding more result numbers. ## Save an experiment-reading card with a clear scope Create a reading card headed this card covers only the selected subsection, followed by source location, field record, omissions and one open question. Clearly separate invented paper-glider values from book data so examples do not enter quotations. After saving or exporting through an available option, reopen it and check that the decimal points in 0.10 and 0.20, millimetre units and not reported labels survive. A later passage can change only the response measure and repeat the workflow, building familiarity with measurement definitions. Date corrections and explain them while retaining a locating phrase. A complete reading card does not mean you reproduced the real experiment. ## Related tutorials and sources [Read arguments in Book Isle: separate claims, evidence and inferential links](); [Read statistical claims in Book Isle: restore denominators, time windows and source tables](). Sources checked on 5 November 2026: the [official Book Isle App Store listing]() and [Apple Lookup]() show version 1.1.3, requiring iOS / iPadOS 15.6 or later. This workflow uses local reading, search, bookmarks and notes for manual checking. Readers record the samples and conclusions; automatic problem solving, experiment analysis or chart recognition is not assumed. Confirm available controls against the installed version and file type. The exercise does not claim step-by-step device testing in this batch. Method reference: [NIST: selecting process variables and responses]() distinguishes input factors from output responses. This workflow checks the completeness of a reading passage, not an experimental-design review. The paper-glider context, values and record fields are invented.